Darko Gjorgjijoski · Skopje, N. Macedonia

I design and ship trustworthy systems.

Software architect and senior engineer with 12+ years taking products from system design through secure, reliable production.

Open to remote software architect and senior/staff engineering roles.

capabilities

The systems view.

Four connected disciplines, applied from the first architecture decision through production operations.
01 / software architecture

Architecture that survives contact with delivery.

I turn product goals and constraints into clear boundaries, interfaces, quality attributes, and evolutionary decisions a team can actually implement.

System boundaries and interfacesTrade-off and risk decisionsIncremental modernization paths
02 / production engineering

Build it, ship it, operate it.

I move across application code, data, delivery, observability, reliability, performance, and cost so architecture stays connected to what actually runs.

Hands-on product deliveryDeployment and observabilityReliability, performance, and cost
03 / security by design

Security in the architecture, not after it.

I model trust boundaries, identity, data exposure, dependencies, and failure modes early, then encode least privilege into the implementation.

Threat models and trust boundariesIdentity and least privilegeData and supply-chain safeguards
04 / AI systems engineering

Useful agents with explicit boundaries.

I design AI-enabled workflows as software systems: tools, context, evaluation, identity, and human control, not a prompt wrapped in a demo.

Agent workflows and tool contractsEvaluation and regression suitesHuman approval and safe failure paths
how I work →
selected work

Architecture with evidence.

Products, platforms, and reference implementations described through constraints, decisions, operations, and outcomes.
all projects and case studies →
blog

Engineering notes.

Direct, practical writing on AI systems, architecture, production engineering, and security.
★ featured
AI Agents Don’t Learn. Stop Saying They Do.
There’s a phrase making the rounds in AI marketing that’s doing real damage: “AI agents continuously learn, execute decisions, and improve automatically.” Sounds impressive. It’s also dangerously misleading. If you take one thing from this post, let it be this: AI agents do not learn. Getting this right matters more than most people realize. What’s […]
2026-03-28 · 3 min ·
all posts →